Skip to content

Alternate page with proper canonical tag: is it a problem?

What Alternate page with proper canonical tag means in Search Console, when the count is fine, and the five template mistakes it can hide, with checks for each.

By Crawlito · 9 min read

In short

  • Google's Page indexing help says an alternate page with proper canonical tag correctly points to an indexed canonical, so there is nothing you need to do.
  • The status only confirms that Google accepted your canonical, not that the canonical you declared was the right one.
  • Google says not to use the first page of a paginated sequence as the canonical, so page 2 and beyond listed here usually means a template mistake.
  • Google asks pages that use hreflang to name a canonical in the same language; if Google accepts a French page's canonical to an English page, the French page is not indexed.
  • The URL Inspection live test shows only the user-declared canonical, because Google decides the selected canonical at indexing time.
On this page

"Alternate page with proper canonical tag" in Search Console means Google found a page that names another URL as its canonical, agreed, and indexed that other URL instead. Google's Page indexing report help says the page "correctly points to the canonical page, which is indexed, so there is nothing you need to do." For tracking parameters, AMP or print versions and other true duplicates, that is the whole story. The status becomes a problem only when the canonical you declared was wrong in the first place, because Google is then accepting a mistake.

This post covers how to tell the two apart, the five patterns that most often hide a real error behind this status, and how to read the declared canonical from raw HTML and HTTP headers. For what a canonical tag is and how to audit one page, see canonical tags and duplicate URLs. Facts below come from Google Search Central and Search Console Help, read on 15 November 2026.

What the status means and what it does not

The Page indexing report sorts URLs Google knows about into indexed and not indexed, with a reason for each. Three of the reasons are about canonicals, and they are easy to confuse:

ReasonWhat happenedUsual action
Alternate page with proper canonical tagThe page names a canonical, Google agreed and indexed the canonicalNone, if the canonical is correct
Duplicate without user-selected canonicalThe page names no canonical, Google picked another URLDeclare a canonical if Google picked the wrong one
Duplicate, Google chose different canonical than userThe page names a canonical, Google indexed a different URLFind the conflicting signal; see when Google chose a different canonical

So "alternate page" is the agreement case. Google followed your hint. What it does not tell you is whether the hint made sense. If a template tells every product page that the homepage is canonical, and Google accepts that for some of them, those product pages show up here as well-behaved alternates while they are missing from search.

Google's canonicalization overview says a search result "usually points to the canonical page." The exceptions it gives are cases where a duplicate suits the user better, such as a mobile page shown to someone on a phone. Otherwise the alternate URL is not the one shown. If the alternate holds content that the canonical does not, that content is effectively out of Google Search. Google's AI features guidance says a page must be indexed to be eligible as a supporting link in AI Overviews or AI Mode, so the same applies there.

When a high count is normal

A large number in this bucket is not a warning in itself. Plenty of sites have more alternates than indexed pages. These sources are expected:

  • Tracking and session parameters: ?utm_source=newsletter, ?gclid=..., ?ref=partner on a page that declares the clean URL.
  • Sort order on a list: ?order=price showing the same products as the default sort.
  • AMP or print versions that point to the main article.
  • Mobile URLs on a separate m. host that point to the desktop page.
  • Host and slash variants that return 200 instead of redirecting, such as a trailing slash version that declares the non-slash URL.

The count can also rise sharply after a marketing campaign, a new affiliate programme or an internal search feature, because Google discovers more parameter URLs. That is not a fault, though if the variants number in the thousands it is worth reading URL parameters and AI search on whether those URLs should be crawlable at all.

Five mistakes this status can hide

Each of these produces an "alternate page with proper canonical tag" entry that looks fine in the report and is actually wrong.

1. A template canonical pointing at the homepage

A theme or SEO plugin sets the canonical from a site-wide setting instead of the current URL, so every page declares https://www.example.com/. Where the pages are clearly different, Google may ignore the hint, and those URLs appear under "Google chose different canonical" instead. Google does not document when it accepts a canonical like this, but pages that are thin or similar, such as tag archives or location pages built from one template, are the ones to check first in this bucket. Google's canonicalization troubleshooting guide names CMS platforms and plugins that misuse canonicalization as a common cause and suggests checking the HTML with browser developer tools.

Sign in the report: example URLs from unrelated sections, all pointing at one URL.

2. Paginated pages canonicalised to page one

Google's pagination guidance is direct: "Don't use the first page of a paginated sequence as the canonical page." It asks you to give each page its own canonical URL. Page 3 of a category lists different products from page 1, so if page 3 points at page 1 and Google accepts it, page 3 is not indexed as a page of its own, even though its list of products is unique. Pagination and infinite scroll covers the linking side.

Sign in the report: ?page=2, /page/3/ or ?p=4 URLs in the examples.

3. Filtered pages canonicalised to the unfiltered list

A filter for "red" or "size 42" shows a subset of the category. Pointing it at the main category is defensible when you do not want filter pages in search. It is a mistake when some filters are pages you want found, such as a brand or a product type with real search demand, because those pages then never get indexed. Decide per filter which ones are landing pages, give those a self-referencing canonical and a crawlable link, and keep the canonical to the parent only for the rest. For sort and filter URLs you do not want indexed at all, Google's pagination page suggests a noindex robots meta tag, or a robots.txt rule to discourage crawling of those URL patterns.

Sign in the report: filter parameters you sell against, such as ?brand= or /category/brand-name/, listed as alternates.

4. Translated pages canonicalised across languages

This is the costliest one for multilingual sites. Google's guide to consolidating duplicate URLs says that if you use hreflang, you should "make sure to specify a canonical page in the same language." Its localized versions page adds that localized versions are only considered duplicates if the main content stays untranslated.

When the French page declares the English page as canonical and Google accepts it, the French page is an alternate and English is the version Google indexes. The hreflang annotations then point at a URL that is not indexed. The fix is for each language version to name itself as canonical and, as the localized versions page requires, list itself as well as every other language version with hreflang:

html
<!-- On https://www.example.com/fr/tarifs -->
<link rel="canonical" href="https://www.example.com/fr/tarifs">
<link rel="alternate" hreflang="en" href="https://www.example.com/en/pricing">
<link rel="alternate" hreflang="fr" href="https://www.example.com/fr/tarifs">
<link rel="alternate" hreflang="x-default" href="https://www.example.com/en/pricing">

Regional versions in one language, such as en-GB and en-US with near-identical text, are the grey area: Google's canonicalization overview lists region variants among the causes of duplicate content, so Google may still cluster them. Hreflang and AI search covers setting those up.

Sign in the report: /fr/, /de/ or country-code hosts listed as alternates of the main-language URL.

5. Parameter variants that change the content

Some parameters do not just track, they change what the page shows: ?variant=blue with a different image and price, ?lang=es, or ?tab=specs that loads different text on the server. If the template treats every parameter as tracking and strips it from the canonical, those distinct pages become alternates. Check each parameter that appears in the examples and ask whether the page content changes with it.

How to check the declared canonical

Search Console tells you that Google accepted a canonical. To see what your server actually declares, read it directly. A page can declare its canonical in two ways you can read from the response: a link element, which Google only accepts in the HTML <head>, and a Link HTTP response header, which Google supports for web search results and which is the usual method for non-HTML files such as PDFs. Sitemaps and redirects are also canonical signals, but they live elsewhere. Google's guide warns against specifying different canonicals for the same page with different techniques, so check both.

The basic curl commands for reading both are in the audit steps of canonical tags and duplicate URLs. For this report, the useful question is narrower: does each example URL name itself or something else? A short loop over a few exported URLs answers it:

bash
for URL in \
  "https://www.example.com/category/shoes?page=3" \
  "https://www.example.com/fr/tarifs"; do
  echo "== $URL"
  # Canonical in the HTML as served, before JavaScript runs
  curl -sL "$URL" | grep -Eio '<link[^>]*rel=["'"'"']?canonical[^>]*>'
  # Canonical in the HTTP response headers
  curl -sIL "$URL" | grep -i '^link:.*canonical'
done

A correct result for the first URL is a canonical naming ?page=3 itself, not the category root, and for the second a canonical naming the French URL. A header canonical looks like this:

http
HTTP/2 200
content-type: text/html; charset=utf-8
link: <https://www.example.com/category/shoes?page=3>; rel="canonical"

Watch for three surprises. A header canonical set by a CDN or server rule that disagrees with the HTML. Two link rel="canonical" elements, often one from the theme and one from a plugin. And a canonical that differs between raw and rendered HTML because JavaScript rewrites it, which JavaScript canonical and noindex tags explains.

Then compare with what Google saw. In the URL Inspection tool, the indexed result shows both User-declared canonical and Google-selected canonical. The live test shows only the user-declared one, because, as Google's URL Inspection help puts it, Google's canonical choice "is determined at indexing time."

A triage routine for the report

The examples table is capped. Search Console Help says it "does not necessarily show all URLs with that issue, and is limited to 1,000 rows." Treat it as a sample:

  1. Open the Page indexing report (Indexing > Pages in the Search Console menu at the time of writing), select the "Alternate page with proper canonical tag" row and export the examples.
  2. Group the URLs by pattern: parameter name, path segment (/page/, /fr/, /tag/), host.
  3. For each group, inspect two or three URLs and note the Google-selected canonical.
  4. Sort each group into "true duplicate" (tracking, sort, AMP, print) or "distinct content" (another page of results, another language, a filter you want found, a variant with different content).
  5. For distinct-content groups, change the template so the page declares itself as canonical, and keep internal links and sitemap entries on that URL.
  6. Switch the report filter from "All known pages" to "All submitted pages". Any URL in your sitemap that appears here is a contradiction: the sitemap says index it, the canonical says index something else.

After you fix a template, Google's troubleshooting guide says it may keep pages in a duplicate cluster for up to two weeks while it re-evaluates them. You can use Request Indexing in the URL Inspection tool, but Google notes it is subject to quotas, so reserve it for your most important URLs.

Frequently asked questions

Do I need to fix "Alternate page with proper canonical tag"?

Not if the canonical is correct. Google's help says there is nothing you need to do; the only work is confirming that the listed URLs really are duplicates of their canonicals.

Why did the count go up after I added canonical tags?

Duplicates that sat under "Duplicate without user-selected canonical" can move to this reason once they declare a canonical that Google accepts. If the canonicals are correct, that shift is the result you wanted.

Should paginated pages show up under this reason?

Not normally. Google says not to use the first page as the canonical of a paginated series, so each page should declare its own URL and stay out of this bucket.

Can I use noindex instead of a canonical on these alternates?

For a true duplicate you want consolidated, Google advises against it, because noindex completely blocks the page from Search. Use noindex only for URLs you do not want indexed at all, such as some sort or filter pages.

Does this status affect AI search tools?

For Google's AI Overviews and AI Mode, the indexed canonical is the version that counts. OpenAI, Anthropic and Perplexity do not document how their crawlers treat canonical tags at the time of writing, so the safe approach is the same: one clear URL per distinct page.

Next step

Canonical template mistakes repeat across every page built from the same template, so one bad pattern can affect hundreds of URLs. The free check shows the declared canonical for a single page as crawlers receive it. To check the pattern across a whole site, with the evidence for each finding, an audit covers up to 500 pages.

Filed under Structure and schema. Topics: canonical, search-console, google, hreflang